NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
Institute of Medicine (US). Preparing for a Changing Healthcare Marketplace: Lessons from the Field. Washington (DC): National Academies Press (US); 1993.
Preparing for a Changing Healthcare Marketplace: Lessons from the Field.
Show detailsStanley B. Jones
Health insurance coverage is one of those topics that makes everybody in the country angry. I learned a long time ago not to tell the person next to me on an airplane that I am involved in private health insurance because everyone is angry about health insurance coverage. Patients feel it is too costly, it does not pay for the services they need, and it involves too much red tape. Employers feel it is too expensive and are worried about its availability. Clinicians are troubled by concerns about how private health insurance shapes medical care; they think of private health insurance as a sort of procrustean bed that has the effect of distorting care. They see it as a form of harassment that determines what they do and how they think about medical problems, as well as who gets paid and how much they get paid. And they are worried about its influence on teaching institutions and the resulting great power it holds over the future of medicine and the training of young professionals.
These days the economists in Washington and assorted health policy theorists have their own gripe: it is with an attitude toward insurance that doctors and patients seem to share. This attitude may be paraphrased as, “Don't worry, insurance will pay for it!” In other words, insurance takes concern with costs out of the transaction between provider and patient.
These gripes arise from deep frustration. The reason for the frustration may be that all of us, despite our sophistication in the insurance industry or in clinical medicine, have an unstated idea about what coverage really should be. And it goes something like this: Insurance should pay for whatever you need to keep well if you are healthy, get well if you are sick, or stay as healthy as possible if you are chronically ill.
Once in a while I do mention on the airplane that I am involved in insurance, and someone will surely say at the end of a discussion of insurance, “Do you remember the days when, when you left the hospital, insurance paid for everything but the television set?” Times are not like that any more. People are troubled about insurance, about what it pays for, and about the extent to which it does not live up to that standard.
WHAT INSURANCE COVERS AND DOES NOT COVER
When people talk about insurance coverage, they really are interested in what insurance pays for. The following is a sample benefit package (it is drawn from the Health Maintenance Organization Act, so it is a particularly liberal benefit package, giving particularly liberal coverage):
- Physician services, including consultant and referral services by a physician;
- Preventive services, including regular checkups, pediatric examinations, immunizations, well-child care, family planning, and infertility treatments;
- Inpatient and outpatient hospital services;
- Emergency health services;
- Diagnostic laboratory and therapeutic radiologic services;
- Home health services;
- Facilities for intermediate and long-term care;
- Vision care;
- Dental services;
- Mental health services;
- Treatment for alcohol or drug abuse;
- Physical medicine and rehabilitative services; and
- Prescription drugs.
Health insurance coverage is initially defined in terms of the types of health care services for which it will pay. Coverage is further defined by criteria these services must meet, such as that those services are performed by a doctor or under the orders of a doctor. An insurance expert once defined health insurance for me by saying, “health insurance pays for what doctors do or order to have done.” We have not traveled very far from that over the past fifty years. There are some areas here and there that are changing, but by and large health insurance payment decisions are based on what doctors do. Coverage is further defined based on where, or in what specific settings, such as hospitals, emergency rooms, and outpatient offices services occur.
Finally, and most importantly, coverage is defined by restricting payment to those services that are “necessary and appropriate for prevention, diagnosis, treatment, and rehabilitation of illness or injury”—a very important phrase in the history of public and private insurance contracting. Incidentally, coverage does not include that much prevention and rehabilitation; it is much more oriented toward diagnosis and treatment of illness and injury.
Of course, some services that might otherwise meet the above definitions are specifically excluded. For example, there is a growing tendency among insurers to exclude autologous bone marrow transplants because they are finding the phrase “necessary and appropriate for diagnosis and treatment” does not protect insurers in court when they think they should be protected. So they specifically exclude certain services if they do not think they are necessary and appropriate. Also, some classes of services are treated differently than others and are excluded from coverage. One example is what insurers persist in calling “nervous and mental conditions” (their phrase not mine). In addition, insurers may exclude preexisting conditions.
For those services that are “covered,” insurance may pay only up to certain limits. The limits may be in terms of the number of visits, the number of days in the hospital, or the total dollar amount of claims paid, for example, $10,000, or $50,000, or $100,000, or $1,000,000 for the lifetime of the policy. Some insurers only pay up to their approved fee schedule for services. So if the doctor charges more, the patient may end up paying the difference. And, of course, insurers only pay after the patient meets deductibles and after coinsurance is subtracted, unless the patient has already met his or her maximum liability for the year, in which case insurers will pay more.
In addition to the above practices, which effectively determine coverage or what an insurer pays, there are implicit or indirect limits that may not be stated. Many older people who are eligible for Medicare have said things to me like, “I was in the hospital and I really wasn't ready to come home, but the doctor (or the hospital) told me that Medicare only covers three days.” This is not true of course, but that is often the way the Diagnostic Related Group (DRG) payment ends up getting interpreted to the patient. This perception has become a limit on coverage of hospital days.
Now, given these coverage definitions, exclusions, and limits, consider the case of someone who has been rendered partly dysfunctional by a series of strokes and is sitting in front of their physician. The physician says, “Here's what we have to do: it's going to involve tests and some expensive imaging; and it's going to involve some hospital care; and there's going to be some physical therapy, maybe long-term care, and rehabilitation services over a period of time.” The patient naturally asks, “Does insurance pay for this, doctor?” Can you imagine trying to answer if you are the physician? Not only do the above coverage criteria make it difficult to answer, but variations exist from insurer to insurer and policy to policy. Unless they happen to work in a one-company town, physicians see a bewildering array of plans. What defines coverage in most of our minds is what insurance pays for—and that always deviates from whatever we have to have done to us to stay well or get well. In this case, neither doctor nor patient will know what is covered until the final bills are received.
INSURANCE MARKETS
Why is it like that? Why does the reality of the insurance mechanism deviate so much from our unstated desire for total coverage? Let me offer a couple of underlying reasons and suggest that if we want to change coverage we have got to get real serious about these underlying reasons. Most attempts to redefine coverages in the past have not done this. People have tried to ignore these underlying reasons—or circumvent them.
The Will of the Buyer
In fact, employers purchase most of the health insurance coverage in this country and they purchase what they want. And what they want is geared toward attracting the employees they need. They do not buy coverage because they believe in good health care. They do not buy it because they are looking after the long-term health of society. They do not even buy it because they are worried about other companies in the community. Employers worry about their own employees and what they need to offer these employees in order to get them to work for them and to keep them working for them.
Thus, what people have by way of coverage varies a lot and it varies largely by labor market. People who are in low-skill labor markets get much less comprehensive coverage than those in high-skill markets. If an employer must attract highly skilled and sought-after engineers, for example, that employer will have a much better health insurance program than the firm down the road that lives largely off of unskilled workers.
Also, employees differ in what they want. Some employee groups are not really interested in some benefits and would much prefer others. Ten years ago, before the cost crunch was as serious as it is now, typically an employer would face a platter of expansion possibilities at the end of his insurance contract year. The platter might include vision, dental, mental health, and other benefits, perhaps even substance abuse coverage. Each benefit would be priced for his employee group. The employer would have the option of picking out this, that, some of this, and some of that, and would often send the insurer back to negotiate an appropriate arrangement. What they bought depended on the biases and desires of their employee groups. If their bias was heavily against what one might call preventive services, then those were excluded. If their bias was against mental health services—and it frequently was—then the employer felt it did not need to offer them. So, the labor market determines what is covered.
The employer also makes trade-off decisions between wages and coverage. The employer can pay employees in either form—higher wages or better coverage. The packaging depends on the market in which the company is competing for employees. The employer also makes trade-offs between personnel costs and prices for goods and services. The company not only competes for employees, it competes to sell products or services in the marketplace. So the employer tries to keep personnel costs high enough to get the needed employees, but low enough to keep prices competitive. Right now, employers are desperate for lower premiums, or, at the least, for premiums that are growing less rapidly from year to year. Right now, also, demand is great among workers for higher wages rather than expansion of other benefits.
A number of problematic insurance coverage and rating activities result from these employer and employee trade-offs and cost pressures. One such problem is experience rating, in which the employer is charged a premium based on the likely future health costs of only their own employees, based on those employees' past health costs. The employer wants to pay for his or her people, but not for anyone else. The insurance industry is getting more and more sophisticated at zeroing in on just one small group and its likely costs, so that the employer pays only for them. Varying coverages are another result. Another is exclusion of preexisting conditions of employees. The insurance industry is getting more and more sophisticated at giving the employer the lowest premium that it can. Higher employee cost-sharing is also a product of this aim, as are cuts in unmanageable benefits. The insurance industry can study the claims experience of an employer and focus on those claims that are costing the most. Mental health coverages in particular have taken a huge beating because of this. They loom large as a percentage of the employer's premium dollar and they often are considered “unmanageable.” It is hard to get control of these services, to be sure that one is only paying for what one wants to pay for.
Cost pressures on employers have also led to some good responses. Health maintenance organizations (HMOs) originally came into the employer market as a response to the need to give young families better coverage at an affordable price.
Insurance as a Limited Product
Another major factor in determining coverage, in addition to the preferences of the buyer, is what I call structural limitations in the insurance product. Insurance is a fundamentally limited product, whether it be from Medicare or public insurance programs or private insurance programs. There are many things that insurance simply cannot do. One of the reasons for this is that insurance is a transaction-based system. It is based on pieces of paper, millions and millions of bills with codes on them that represent services to subscribers. Insurers process bills, and their entire structure is built on the flow of bills. How it got that way for most of the country could perhaps make for an interesting article. Maybe it grew out of the industry's beginnings as a financial protection institution. Insurance did not set out to provide health care, it set out to protect people against the cost of health care. In any case, because of this flow of bills, insurers have huge investments in computers, and a large proportion of an insurer's employees are wrapped up in this processing activity. The way insurers think about their role is shared by this history. The upshot of it of course is computers and software and telephones. One may also attribute the sense of harassment and distance that most clinicians feel from insurance companies to these computers and telephones.
Another major limit, and perhaps a more fundamental one, is that in order to predict costs and set a premium, insurers have to quantify the use of services that are expected from a certain coverage. They must be able to count the providers or service sites and describe discrete service units. These factors are all interrelated. Often it is difficult to describe the service unit with enough precision for someone who is reviewing a bill to be able to differentiate it from another source, be assured that it was needed, and put a reasonable price on it. If a service is hard to describe for the above purposes, insurers can sometimes resort to limiting the providers who will offer it. In other words, if the service cannot be described, then the insurer can at least make sure that a doctor provides it. If insurers say that a doctor can only provide the service in a hospital, that limits the coverage even more. One of the upshots of all of this need to describe services so as to predict how many will be covered is that insurers avoid “soft” services, those that are difficult to define.
Another limit is the need to manage care. Here the question for the insurer is whether a service is necessary and appropriate for the diagnosis and treatment of illness or injury. The treatment may be worthwhile, but nevertheless not necessary and appropriate. Now, bear in mind that this is determined through pieces of paper, at “arms length,” so to speak. In fact, there is something of a religion among insurance companies about being at arms length from the providers. This means the piece of paper has to be adequate to tell whether a service is necessary and appropriate for diagnosis and treatment. This leads to a cops and robbers game wherein the physician tries to look out for the patient and checks those boxes on the form that look most helpful to the patient, none of which exactly describe what was done, and the insurer knows what is going on and tries to stop it and find a way to trim back costs. Each side gets tougher and tougher.
The industry uses averages and percentages of the mean in physician practice patterns in order to decide what is “necessary and appropriate” and what to pay for. There is hope that more sophisticated protocols are being developed by clinicians as a substitute for this kind of review. Pre-admission certification is another of the ways in which the industry protects itself. In this way the insurer argues about whether the service is necessary before rather than after it is performed.
This structural limit also explains why insurers tend to start independent practice association (IPA) and preferred provider organization (PPO) panels rather than group and staff models. An IPA or PPO still handles paper, it is still an arms-length arrangement, and it is still transaction based—and an insurer can understand this. The idea of sitting down around a table with physicians and hospital administrators is a very difficult concept for people in the insurance culture.
Another of the structural limits of the insurance industry is the “moral hazard problem.” There is some moral hazard for everybody involved in the insurance process. For the insurance buyer, there is the temptation not to insure until you get sick. Young people especially fall into this category. So when they get sick and go shopping for insurance, they find the insurance company has excluded preexisting conditions, or that the insurance company declines to sell them insurance at all (to keep from insuring only sick people).
For the provider there is the moral hazard of going ahead and performing a procedure since it is covered, even if it is marginal in value, or of coding a service in a way that assures it will be paid for, in order to help the patient out a little. The insurer responds to these provider practices by requiring precertification, tighter fee schedules, and what I call current procedural terminology (CPT) code wars. In these code wars the insurer studies how CPT codes are being measured and refines payment rules. The physicians on the other side are looking for ways to use the codes, and to create new codes, that will enable them to maintain their revenues. It is astounding that these codes are still controlled by organized medicine, given the reality and costliness of this war. If other tactics fail, the insurer's last response is participation.
The patient has his own version of this moral hazard, and that is to ask for a procedure simply because it is covered. The answer to that from the insurer is cost-sharing, putting some of the burden on the individual so that they hesitate to ask for treatment.
Finally, as if there were not already enough structural limits on insurance, the need for insurers to offer competitive premiums and products poses one more. Insurance is a business. Employers can have their pick of insurers. It is a competitive industry characterized by strong fads and trends in the market. Often too little research backs up these trends. Over a given two- or three-year period, big employers all over the country do the same things. They add certain benefits, they add certain managed care practices. Often, the similarity of practices is furthered by benefit consulting companies that advise employers. When these consulting companies are asked to show data to prove that their measures save money, it is incredible how thin their data are. Sometimes they are nonexistent. Nevertheless, belief in certain measures is widespread. Insurance is an industry that is truly racked by fads, and very little is being invested in research and development.
Also, unfortunately, in insurance one can achieve a lower premium more quickly and efficiently by sophisticated rating, underwriting, and risk selection than by managed care. It is still the case that if an insurer wants to sell a smallish employer a good policy, that insurer is most likely to win the competition by figuring out how to rate better than another insurer or underwrite in a more sophisticated way, perhaps by offering a package that excludes certain conditions or certain employees. That is still the winning approach. In fact, it is frightening that ten years from now we may look back at the 1980s and 1990s, not as the era when managed care mushroomed among insurers, but as the era when insurers became more adept at rating and underwriting.
One of the sad upshots of the limits of insurance and particularly risk selection, is that one population is particularly poorly served. People who are chronically ill may in fact find a better clinical package and better care with an HMO—even a group or staff model HMO—if they only know it is available to them. But in fact some HMOs know that if they advertise that they can take care of, for example, AIDS patients in a way that was more satisfactory to the patient and less costly, then they would end up getting all the AIDS patients. And no matter how efficient these HMOs are, they cannot be competitive as an insurer if a highly disproportionate share of sick people have chosen them. The smart thing to do is not to advertise these benefits, but to let people stay with their original insurer unless they find their way to the HMO. Then, when they get sick, they can be congratulated on their good luck and their care can be covered. We are all losing innovation and efficiency in our system because of these structural limits of insurance.
Now, before going on, I have to say that these are generalizations. We could talk about how some insurers and, especially, how some group and staff model HMOs and some clinics, hospitals, and physician groups are doing some truly innovative and interesting things. However, we hear so much about these innovations, we might get the impression that they represent some substantial portion of the industry. Alas, they do not. They represent a small part of what is going on out there; it is a hopeful part but a very small part.
IMPROVING COVERAGE
So, given the limits detailed above, if you want to improve coverage, what can you do? If your intention is to improve coverage, what kind of government improvements might help, and what kind of government changes?
First, you have to decide whether it is better to reform, to rebuild, or to replace the employer as the payer for, and purchaser of, coverage. Those are really the three options: reform, rebuild, or replace. If you think about it, it is incredible that what is covered for an employee is so dependent on accidents of employment: whether there happens to be a union negotiating benefits, what the employee's skill level is, which employer the employees happen to go with, and all sorts of other accidents that wait out there in the job market. These accidents are what determine most coverage. In order to influence that determination, we need a more careful definition of the social good of different kinds of health services and of the quality of various kinds of health services. That is the unfinished agenda.
Ten or twelve years ago I was on a panel at the Institute of Medicine, and the subject was mental health. At the time I was frustrated because I was looking for some relevance to insurance of our discussions on mental health services. And I asked this group of able and renowned clinicians if maybe we could define a core benefit made up of those services that do good, that really work, and that help people. Now, if you are an employer, and do not have a lot of money, and are trying to decide between vision care and dental care and other items, and you could be talked into providing some mental health coverage, which services would you start with? Or, if you were willing to spend more, is there a broader set of services you might tackle? Well, the panel could not agree. Not only could we not agree on what should be on the inexpensive list, we could not even agree on the efficaciousness of possible items for the list.
If we want to improve insurance coverage, I think the burden is on clinicians and researchers to tackle that tough problem. The discussion of outcomes research is a great framework or culture in which to be talking about this, but the burden to demonstrate its social good will finally be placed on those who feel strongly about coverage.
Next, having decided whether to reform, rebuild, or replace the employer as buyer of health insurance, we have to decide whether we want to reform, rebuild, or replace private health insurance. Those who are reform minded look at that list of limits in health insurance and see the possibility that the present system can work. They argue that the problem is that we have too much of this and too little of that and this service is not well defined and that procedure is clumsy, and so on. The burden to reformers is to come up with better definitions of services so these services can be counted, so they can be costed out, and so an actuary can decide the allowable number and cost of these services and therefore, ultimately, the premium. Definitions are also needed, for providers and service sites, of the term “necessary and appropriate” so that reviewers of claims can review them better. We also need a better definition of “experimental.” Incidentally, it does not matter whether we are talking about a public or a private system. Medicare has to reform in this way as well. So, for advocates of a single-payer system who favor the Medicare insurance model, these same questions must be answered.
Those who are rebuilders and who think we can and should move into vertically integrated plans face a truly huge change, a rebuilding of the entire insurance industry. I think it is a greater change than General Motors changing its auto lines, much greater. Advocates of rebuilding are talking about recreating and restructuring the foundations of an industry. If rebuilding is the plan, then providers must be involved. The “arms-length” issue has been destructive from both sides. Providers are safely removed from insurance and in fact have been made a lot wealthier by the existence of health insurance in this country. Rebuilding insurance in order to improve coverage means moving providers into integrated systems so that they share accountability and responsibility for managing both clinical aspects of care and cost aspects of care. Ways must be found to do that without compromising professional ethics and values.
Some flexibility in coverage, as well as room for innovation and greater responsiveness to patient needs, may arise from this. Because, in fact, if the provider and the insurer could be jointly responsible, with payments made in terms of capitation amounts and budgets, then flexibility could be achieved in all the definitions of what services will be deemed necessary and appropriate, by what providers, and at what sites.
There are those who wish to replace private insurance with government. Many people feel this way these days, but frankly, they will end up limited by the same mistakes. They will be faced by the same structural limits, but even more so. If we want the government to be the insurer, we have to work harder at defining the services, the providers, the sites and the term “necessary and appropriate.”
Lastly and most importantly, we are not going to improve coverage until we contain the rising costs of health care. The fact that costs have gotten out of hand has stopped the growth in benefits and experimentation in the last ten years. And it has encouraged all the definition sharpening, the exclusions, and ultimately the reductions in what insurance pays for. The fact is, we are moving farther and farther from our basic goal of covering what is needed to get us well or keep us well. And the reason is costs. There are two formulae to controlling costs. One is to rebuild insurers and to restructure employment-based coverage so as to move our system to vertically integrated health plans; the other is to regulate providers directly and directly set rates. At present, we are in the midst of the debate between these approaches. Ultimately, neither will improve coverages unless they contain costs.
- The Evolution of Private Health Insurance: Past Issues, Future Challenges - Prep...The Evolution of Private Health Insurance: Past Issues, Future Challenges - Preparing for a Changing Healthcare Marketplace
Your browsing activity is empty.
Activity recording is turned off.
See more...